> I think you should probably do this differently. Basically the RRD > file format is a number of RRAs, each RRA being built from a number > of rows, each row being built from a number of CDPs. What you need > to do is:
After fighting trying to learn my way around in the code, I finally gave up and agreed. I'm now about 80% done with doing it entirely in perl, after a translation of rrd_*.h into a perl script. Probably overboard, but I wanted the entire RRD file to be understood from perl for future use. At this time I can read the entire RRD header, and am coding the part that copies the header, learns defaults, and writes a modified header. I hope to be done in a couple hours and on to actual testing. The one thing I did have to do, that I kinda disagree with but am doing it out of neccessity, is making assumptions on how things are byte-aligned. Specifically, all of the double-words are always aligned on a 8 byte boundary. This assumption may not be right on other platforms, but I've had to make it be enough for now. The cookie check, if it succeeds, is good enough to say that the assumptions are safe though. Thanks for the feedback thusfar. -- Unsubscribe mailto:[EMAIL PROTECTED] Help mailto:[EMAIL PROTECTED] Archive http://www.ee.ethz.ch/~slist/rrd-users WebAdmin http://www.ee.ethz.ch/~slist/lsg2.cgi
